Chinese Journal of Chemical Physics  2018, Vol. 31 Issue (5): 667-672

#### The article information

Shuai Qin, Li-wei Hui, Li-hua Yang, Ming-ming Ma

Solvent-Triggered Self-Folding of Hydrogel Sheets

Chinese Journal of Chemical Physics, 2018, 31(5): 667-672

http://dx.doi.org/10.1063/1674-0068/31/cjcp1803025

### Article history

Received on: March 22, 2018
Accepted on: April 27, 2018
Solvent-Triggered Self-Folding of Hydrogel Sheets
Shuai Qin, Li-wei Hui, Li-hua Yang, Ming-ming Ma
Dated: Received on March 22, 2018; Accepted on April 27, 2018
CAS Key Laboratory of Soft Matter Chemistry, School of Chemistry and Materials Science, Universityof Science and Technology of China, Hefei 230026, China
*Author to whom correspondence should be addressed. Li-hua Yang, E-mail:lhyang@ustc.edu.cn; Ming-ming Ma, E-mail:mma@ustc.edu.cn
Abstract: Intense investigations have been attracted to the development of materials which can reconfigure into 3D structures in response to external stimuli. Herein we report on the design and self-folding behaviors of hydrogels composed of poly(ethylene glycol) methyl ether methacrylate (OEGMA) and 2-(2-methoxyethoxy) ethyl methacrylate (MEO$_{2}$MA). Upon immersion into a variety of solvents at room temperature, the resulting P(MEO$_{2}$MA-co-OEGMA) hydrogel sheets self-fold into 3D morphologies, and the observed transformation in shape is reversible. We further show that composition of the gel, gaseous environment, and preparation procedure play important roles in the self-folding behavior of the resulting hydrogels. This work provides a facile approach for fabricating self-folding hydrogels.
Key words: Self-folding    Solvent responsive    Shape transformation    Hydrogel
Ⅰ. INTRODUCTION

Mechanically active, self-shaping materials such as hydrogels, which undergo prescribed shape transformations in response to external stimulus, can mimic the sensing and responsive mechanisms found in nature [1]. Particular interest has been directed toward the development of intelligent soft materials that reconfigure into 3D structures in response to specific external triggers [2-5]. The differential swelling leads to enhanced tendency of internal stresses, resulting in reversible transformations between three-dimensional shape and two-dimensional sheets. Additionally, previously explored stimuli generally result in different shape transformation mechanisms, determined by the magnitude of the applied stimulus. A multitude of external stimuli can induce hydrogels to undergo large change in volume, which alters the polymer-solvent interactions [6]. The stimulus-induced change in volume can be harnessed to create shape transforming materials by approaches such as spatially varying the degree of cross-linking [7, 8], adopting a multilayer design [9-11], or attaching the hydrogel to a rigid surface so as to impose a mechanical boundary constraint [12], having often been introduced to achieve stable differential swelling. Appropriate adaptations of these methods could point to pathways for generating complex 3D structures applications in biotechnology [13], sensor [14], microfluidic devices [15, 16] and actuator [17, 18]. Sometimes these gel systems require multiple manufacturing operation steps.

Copolymers P(MEO$_{2}$MA-co-OEGMA) can be potentially connected easily to a wide variety of synthetic polymers, biological structures and inorganic surfaces, acting as a promising replacement of conventional PNIPAM for applications and more universal for building any kind of thermo-sensitive materials [19, 20]. Herein, we report an approach to generate 3D hydrogel structures. The technique involves polymerization of monomers into P(MEO$_{2}$MA-co-OEGMA) hydrogel. The general idea is to use a polymer which is able to fold in response to multi-solvents. In our work, we used radical polymerization to prepare P(MEO$_{2}$MA-co-OEGMA) hydrogel. Afterwards the polymer was cured in a groove, which is a simple way to finish this process. We further showed that by controlling gel composition, gaseous environment, and forming process, the P(MEO$_{2}$MA-co-OEGMA) gel can self-assemble into 3D morphologies. The crosslinked hydrogels swell and non-uniformly expand in diverse solvents. The swelling of a hydrogel enables the hydrogel to roll up and eventually form tubes. The observed transformation in shape does not require a difference in structure or composition across the thickness of the hydrogel sheet.

Ⅱ. EXPERIMENTS A. Materials

Poly (ethylene glycol) methyl ether methacrylate (OEGMA, Aldrich, M$_{\textrm{n}}$=475 g/mol), 2-(2-methoxyethoxy) ethyl methacrylate (MEO$_{2}$MA, 9dingchem, M$_{\textrm{n}}$=188 g/mol), ammonium persulfate (APS, sinopharm chemical reagent Co., Ltd), tetramethylethylenediamine (TEMED, Aldrich).

B. Synthesis of hydrogel in air

1.4412 mg MEO2MA and 0.9106 mg OEGMA (molar ratio=80:20) were mixed with 588 $\rm{\mu }$L of 50 mg/mL ammonium persulfate aqueous in a 10 mL centrifuge cup via vortex for 30 s, 11.76 $\rm{\mu }$L TEMED was then added into it via vortex for 30 s, after that we dump the liquid into the groove (2 cm$\times$5 cm$\times$1 cm), and waited for the mixture curing overnight. MEO$_{2}$MA:OEGMA=50:50 and OEGMA hydrogel kept the same total mole number of OEGMA and MEO$_{2}$MA and other parameters didn't change.

C. Synthesis of hydrogel in N$_{2}$

Before the hydrogel was put into the groove, the groove was placed in a bottle inlet N$_{2}$ for 15 min, then 1.4412 mg MEO$_{2}$MA and 0.9106 mg OEGMA were mixed with 588 $\rm{\mu }$L of 50 mg/mL ammonium persulfate aqueous in a 10 mL centrifuge cup via vortex for 30 s, then the liquid was inlet N$_{2}$ for 10 min, and 11.76 $\rm{\mu }$L TEMED was added into it via vortex for 30 s, after that we injected the liquid into the groove (2 cm$\times$5 cm$\times$1 cm) and aerated N$_{2}$ for another 15 min, then sealed the outlet of the bottle, waited for the mixture curing overnight.

D. Characterization of hydrogel

The morphology of hydrogel was examined by using an environmental scanning electron microscope (Philips XL30 ESEM-TMP) at an acceleration voltage of 10 kV. The FTIR spectra were characterized on a Thermo Scientific OMNIC spectroscopy (Nicolet 6700) from 4000 cm$^{-1}$ to 400 cm$^{-1}$ at room temperature. The UV-Vis spectra were measured on Shimadzu UV-3600 Plus from 250 nm to 500 nm at room temperature. There are three groups of liquid including 1 mL 40 mg/mL KI with 500 $\rm{\mu }$L pure water, 1 mL of 40 mg/mL KI with 500 $\rm{\mu }$L OEGMA and 1 mL pure water with 500 $\rm{\mu }$L OEGMA.

E. Mechanical tests of hydrogel

For tensile measurement, the hydrogels were made into strips of 1.5 cm in width, 7 cm in length (the effective length was 5 cm) and 2 mm in thickness. Tensile measurements were performed by uniaxially stretching the strips of hydrogels at a strain rate of 50 mm/min.

Ⅲ. RESULTS AND DISCUSSION

The synthetic method of hydrogel was free radical polymerization as FIG. 1(a) shows. The gelation of P(MEO$_{2}$MA-co-OEGMA) hydrogel utilized autoxidation process of OEGMA. If we put P(MEO$_{2}$MA-co-OEGMA) hydrogel sheet into pure water, it bended in several seconds as shown in FIG. 1(b). Differential swelling resulting from internal stresses induced shape transformations of hydrogel sheet. In addition, polymerization rate largely relied on the availability and abundance of free radicals in the reaction system. Gaseous environment made extent of reaction in surface lower than inner, consequently degree of crosslinking of the surface of hydrogel was slightly lower, making hydrogel frizzle easier.

 FIG. 1 (a) The polymerization and crosslink formation of poly(ethylene glycol) methyl ether methacrylate (OEGMA) and 2-(2-methoxyethoxy) ethyl methacrylate (MEO$_{2}$MA), which lead to the formation of the as-reported hydrogels. (b) Schematic illustration on the sheet self-folding of a resulting hydrogel upon immersion into a solvent and shape recovery after being taken out from the solvent.

According to the SEM image of MEO$_{2}$MA: OEGMA=80:20 hydrogel, there are abundant wrinkles on the surface of hydrogel and no obvious porous structure is observed (FIG. 2(a)). We synthesized three kinds of hydrogels including MEO$_{2}$MA:OEGMA=80:20, MEO$_{2}$MA:OEGMA=50:50 and OEGMA alone. The spectra of these three hydrogels are similar. With the increase of OEGMA amount, we find the peak at the wavenumber of 1125 cm$^{-1}$ becomes more apparent, and the peak corresponds to CH$_{2}$$-$O$-$CH$_{2}$ of OEGMA ether chain. The other peaks at 2850, 1740, and 1450 cm$^{-1}$ correspond to $-$CH$_{2}$, C$=$O, and O$-$CH$_{3}$ in hydrogel, respectively (FIG. 2(b)).

 FIG. 2 (a) Scanning electron microscopy image of hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20. (Inset) Photography of the hydrogel in a test tube. (b) Infrared spectroscopy of (blue line) hydrogel composed of OEGMA alone (i.e. O(100)), (red line) hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=50:50 (i.e. M(50):O(50)), and (black line) hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20 (i.e. M(80):O(20)). (c) Relationships of stress versus strain for (blue line) hydrogel composed of OEGMA alone (i.e. O(100)), (red line) hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=50:50 (i.e. M(50):O(50)), and (black line) hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20 (i.e. M(80):O(20)). (d) Ultraviolet-visible spectra of (black line) KI, (red line) OEGMA, and (blue line) the mixture of KI and OEGMA.

According to the stress strain curve of three hydrogels (FIG. 2(c)), MEO$_{2}$MA:OEGMA=50:50 hydrogel (elongation at break strain 190%) is close to MEO$_{2}$MA:OEGMA=80:20 hydrogel (elongation at break strain 208%) in elongation, but with the increase of OEGMA the Young modulus of hydrogel increases. MEO$_{2}$MA:OEGMA=50:50 hydrogel (Young modulus 17.4 kPa) have better mechanical property than MEO$_{2}$MA:OEGMA=80:20 hydrogel (Young modulus 5.3 kPa). Compared to MEO$_{2}$MA:OEGMA=50:50 hydrogel (Young modulus 17.4 kPa), OEGMA hydrogel (Young modulus 16.1 kPa) has approximate Young modulus but elongation at break decreases obviously. When the elongation reaches only 75%, the hydrogel becomes crisp. According to the results of different hydrogels, we found that with OEGMA content increasing, the mechanical strength of hydrogel improves, but high levels of OEGMA make hydrogel become crisp. Mechanical properties of these three kinds of hydrogel is closely related to its deformation behavior.

Crosslinking agent plays an important role in hydrogel formation, in this work we found that the system without crosslinking agent can also finish this process. To investigate the reason of hydrogel formation, we made three groups mixture with different components. The mixture of OEGMA, APS, and TEMED could form hydrogel after several hours; the mixture containing MEO$_{2}$MA, APS, and TEMED still maintained as liquid for a very long time; the group containing OEGAM and APS finished gelation process after 72 h. It needed longer time, because the redox reaction became slowly, the gelation process was slow and the hydrogel absorbed water, so we fabricated a swelling hydrogel different from other groups. We consider that OEGMA is an important factor in this hydrogel formation reaction. Hydrogel formed without crosslink agent, we suppose there is an autoxidation process [21] in the long ether chain of OEGMA.

In order to verify it, we used the system of KI aqueous [22]. If oxide existed in OEGMA, we could find it by this reagent. When OEGMA was added into the KI solution, we could find a new peak near 350 nm, which corresponds to the I$_{2}$ in the KI solution, but KI aqueous and OEGMA in water didn't have this peak (FIG. 2(d)). According to this, we could affirm that oxide in the OEGMA transfers KI to I$_{2}$. The long ether chain of OEGMA provided site for reaction, and autoxidation process spurred crosslinking occurred. According to the UV-Vis results, we calculated the number of reaction sites in ether chain. The concentration of reaction site is near 4.42$\times$10$^{-8}$ mol/mL in OEGMA, and one reaction site exists in per 25 OEGMA side chains, in which the polymer completes the curing process. According to the number of reaction site, we estimate the hydrogel will be fragile, which is also proven by tensile measurement.

Monomers with long ether chain make hydrogel absorb different kinds of solvents, which helps the shape transformation of hydrogel in multi-solvents. As FIG. 3(a) shows, if we put the hydrogel into the pure water, it will start to bend after several seconds. Finally, it will become a tube by its long axis due to the curve of the hydrogel. After water evaporation, the hydrogel will recover flat, namely it is solvent-triggered reversible shape transformation hydrogel.

 FIG. 3 (a) Hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20 undergoes shape transformation upon being immersed into water and shape recovery after water evaporation. (b$-$d) At 1 min after immersed into water, the shape formed by (b) hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20, (c) hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=50:50, and (d) hydrogel composed of OEGMA alone. Scale bar=1cm.

We also noted that composition of hydrogel will affected the shape transform process, and we made three different hydrogels with composite MEO$_{2}$MA: OEGMA=80:20, MEO$_{2}$MA:OEGMA=50:50, and OEGMA alone. They were all made by glass mould, after solidification we put them into the pure water. Three hydrogels can bend with different speed when immersed into the water. As FIG. 3(b-d) show, with the percent of MEO$_{2}$MA increasing, bend speed of hydrogel becomes faster. When it comes to the system with less OEGMA, the crosslinking point will reduce, and more soft hydrogels have fast speed in self-folding process.

Because the polymerization rate largely relies on the availability and abundance of free radicals in the reaction system, this process is susceptible to the existence of free-radical inhibitors, molecules that can convert free radicals into much less reactive or even nonreactive counterparts [23]. O$_{2}$ as a common inhibitor can turn free radicals into radicals that have insufficient reactivity to continue polymer chain growth, and therefore slows down the gelation process [24]. Therefore, special treatments to control dissolved O$_{2}$ have often been performed in preparing hydrogel [25]. In this work, we take advantage of the detrimental effect of O$_{2}$ on polymerization of P(MEO$_{2}$MA-co-OEGMA) hydrogel and turn it into an approach to manipulate the gelation process. If the hydrogel in the environment has less oxygen, the reaction can finish more thoroughly. It will entitle hydrogel with slow bend speed as FIG. 4(b) shows, but on the other hand, in the environment with less oxygen, we can make a thinner hydrogel, so we may find more obvious phenomenon in some conditions.

 FIG. 4 The shape transformation in water as a function of time for hydrogel with same composition (MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20) but prepared in different mould and under different atmosphere: (a) in glass mould and under air, (b) in glass mould and under N$_{2}$, (c) in teflon mould and under air. Scale bar=1cm.

We further demonstrate another factor that can influence shape transformation which is associated with forming process. Shape transformation of hydrogel focuses on long axis, when we changed the shape of the hydrogel, it showed the same appearance. We found because of surface tension, hydrogel in a glass mould is not complete flat, and the edge of the hydrogel is curve, so we think the curve induce the bend of the hydrogel. We changed the mould material from glass to PTFE, and the two moulds have the same size. In PTFE mould, hydrogel had a curve in a small angle, then we put the hydrogel into the pure water, and we found that the same shape hydrogel made by different moulds have different bend speed in water as FIG. 4(c) shows. In addition, we make a hydrogel by a square glass mould, after immersed in water we did not find hydrogel bent along one side, until 30 min we only found hydrogel become transparent because of water absorption. We found when curve exists around the border, hydrogel tended to bend toward center, but offset of stresses made hydrogel kept its shape. According to these phenomena, we think curve in hydrogel border plays an important role in this bend process, if we change the shape of hydrogel edge, we may change the shape transformation mode, so we can make the hydrogel with complex shape and accomplish self-folding process after being submerged in pure water, and such transition may have potential application in soft devices.

Next, we characterized the bend speed of MEO$_{2}$MA: OEGMA=80:20 hydrogel in different solvents. As FIG. 5(a-d) show, in different solution, hydrogels showed different bend speed. In ethanol, hydrogels had slow bend speed. When hydrogels were put into chloroform, we found hydrogels bent rapidly and finished this process after 15 s. Hydrogels swelled and floated in the solvent in the end. In other two solutions we found no big difference in bend speed, and while bending we didn't find obvious swelling. After shape transformation, hydrogel becomes transparent because of solvent absorption. According to the phenomenon of self-folding of hydrogel in different solution, we found that the bend of hydrogel in solution is in association with swelling of hydrogel. In chloroform hydrogel had the fastest swelling speed, and shape transformation process was finished as soon as possible. According to phenomenon we observed, hydrogel can swell in different kinds of solvent. It can be regarded as an amphipathic material, because two monomers have long ether sections, and it improves interaction between hydrogel and solvent (FIG. 5(e)). If we increase the content of OEGMA, it has similar solvent absorb behavior in these four kinds of solvent.

 FIG. 5 (a$-$d) The shapes of hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20 at 1 min after immersed into (a) water, (b) ethyl acetate, (c) ethanol, and (d) chloroform. Scale bar=1 cm. (e) Relationship of relative volume of solvent absorbed versus immersion time for hydrogel composed of MEO$_{2}$MA and OEGMA at molar ratio of MEO$_{2}$MA:OEGMA=80:20 in (black) ethanol, (red) water, (blue) ethyl acetate, (green) chloroform.
Ⅳ. CONCLUSION

In summary, we demonstrate an approach to generate multi-solvents responsive self-folding P(MEO$_{2}$MA-co-OEGMA) hydrogel via free radical reaction, and long chain monomer OEGMA plays an important role in hydrogel formation. With long side chain existence, the hydrogels have better polymer-solvent interactions in multi-solvents.

It demonstrates the possibilities that generating heterogeneous phase system using homogeneous reagent. Differential growth of hydrogel can be effectively introduced when gaseous environment are incorporated into the polymerization system. Inhomogenous hydrogel make their shape transformation behavior in solvent easier. Hydrogel composition and solvent environment also influence the speed of shape transformation. If we change the curve of the hydrogel surface, we can control the bend direction of hydrogel, and the approach can potentially be used to hydrogel component. The hydrogel can be assembled into 3D scaffolds which are likely to be applicable to a wide variety of fields, such as engineering, polymer science, soft robotics, and flexible electronics.

Ⅴ. ACKNOWLEDGMENTS

This work was supported by the Youth Innovation Promotion Association of Chinese Academy of Sciences.

Reference
 [1] R. M. Erb, J. S. Sander, R. Grisch, and A. R. Studart, Nat. Commun. 4 , 1712 (2013). DOI:10.1038/ncomms2666 [2] Z. J. Wei, Z. Jia, J. Athas, C. Y. Wang, S. R. Raghavan, T. Li, and Z. H. Nie, Soft Matter 10 , 8157 (2014). DOI:10.1039/C4SM01299B [3] P. D. Topham, J. R. Howse, C. J. Crook, S. P. Armes, R. A. L. Jones, and A. J. Ryan, Macromolecules 40 , 4393 (2007). DOI:10.1021/ma071041j [4] T. S. Shim, S. H. Kim, C. J. Heo, H. C. Jeon, and S. M. Yang, Angew. Chem. Int. Ed. 51 , 1420 (2012). DOI:10.1002/anie.201106723 [5] R. Kempaiah, and Z. H. Nie, J. Mater. Chem. B 2 , 2357 (2014). DOI:10.1039/C3TB21462A [6] E. M. White, J. Yatvin, J. B. Grubbs Ⅲ, J. A. Bilbrey, and J. Locklin, J. Polym. Sci. B: Polym. Phys. 51 , 1084 (2013). DOI:10.1002/polb.23312 [7] H. Thérien-Aubin, M. Moshe, E. Sharon, and E. Kumacheva, Soft Matter 11 , 4600 (2015). DOI:10.1039/C5SM00561B [8] Y. Klein, E. Efrati, and E. Sharon, Science 315 , 1116 (2007). DOI:10.1126/science.1135994 [9] Z. L. Wu, M. Moshe, J. Greener, H. Therien-Aubin, Z. H. Nie, E. Sharon, and E. Kumacheva, Nat. Commun. 4 , 1586 (2013). DOI:10.1038/ncomms2549 [10] H. Thérien-Aubin, Z. L. Wu, Z. H. Nie, and E. Kumacheva, J. Am. Chem. Soc. 135 , 4834 (2013). DOI:10.1021/ja400518c [11] P. Techawanitchai, M. Ebara, N. Idota, T. A. Asoh, A. Kikuchi, and T. Aoyagi, Soft Matter 8 , 2844 (2012). DOI:10.1039/c2sm07277g [12] H. Lee, J. P. Zhang, H. Q. Jiang, and N. X. Fang, Phys. Rev. Lett. 108 , 214304 (2012). DOI:10.1103/PhysRevLett.108.214304 [13] L. Liu, N. Wang, Y. J. Han, Y. M. Li, and W. G. Liu, Macromol. Rapid Commun. 35 , 344 (2014). DOI:10.1002/marc.v35.3 [14] J. Z. Hilt, A. K. Gupta, R. Bashir, and N. A. Peppas, Biomed. Microdevices 5 , 177 (2003). DOI:10.1023/A:1025786023595 [15] D. J. Beebe, J. S. Moore, J. M. Bauer, Q. Yu, R. H. Liu, C. Devadoss, and B. H. Jo, Nature 404 , 588 (2000). DOI:10.1038/35007047 [16] G. H. Kwon, Y. Y. Choi, J. Y. Park, D. H. Woo, K. B. Lee, J. H. Kim, and S. H. Lee, Lab Chip 10 , 1604 (2010). DOI:10.1039/b926443d [17] E. Palleau, D. Morales, M. D. Dickey, and O. D. Velev, Nat. Commun. 4 , 2257 (2013). DOI:10.1038/ncomms3257 [18] B. Kaehr, and J. B. Shear, Proc. Natl. Acad. Sci. USA 105 , 8850 (2008). DOI:10.1073/pnas.0709571105 [19] J. F. Lutz, K. Weichenhan, O. Akdemir, and A. Hoth, Macromolecules 40 , 2503 (2007). DOI:10.1021/ma062925q [20] Z. Q. Meng, F. Wei, R. H. Wang, M. G. Xia, Z. G. Chen, H. P. Wang, and M. F. Zhu, Adv. Mater. 28 , 245 (2016). DOI:10.1002/adma.201502669 [21] Z. Kerem, W. L. Bao, and K. E. Hammel, Proc. Natl. Acad. Sci. USA 95 , 10373 (1998). DOI:10.1073/pnas.95.18.10373 [22] N. V. Klassen, D. Marchington, and H. C. E. McGowan, Anal. Chem. 66 , 2921 (1994). DOI:10.1021/ac00090a020 [23] A. Giz, H. Çatalgil-Giz, A. Alb, J. L. Brousseau, and W. F. Reed, Macromolecules 34 , 1180 (2001). DOI:10.1021/ma000815s [24] C. Decker, Macromolecules 18 , 1241 (1985). DOI:10.1021/ma00148a034 [25] C. J. Huang, D. Quinn, S. Suresh, and K. J. Hsia, Proc. Natl. Acad. Sci. USA 115 , 70 (2018). DOI:10.1073/pnas.1717912115